Update app.py

app.py

@@ -1,16 +1,28 @@
+import os
 import torch
 from fastapi import FastAPI, Request
 from transformers import AutoTokenizer, AutoModelForCausalLM
 from peft import PeftModel
 
+# 🔐 HF_TOKEN değişkenini oku
+hf_token = os.getenv("HF_TOKEN")
+
+# 🔧 Model bilgisi
 base_model_id = "google/gemma-1.1-2b-it"
 lora_model_id = "programci48/heytak-lora-v1"
 
-tokenizer = AutoTokenizer.from_pretrained(base_model_id)
-base_model = AutoModelForCausalLM.from_pretrained(base_model_id, torch_dtype=torch.float32, device_map=None)
-model = PeftModel.from_pretrained(base_model, lora_model_id)
+# 📦 Model yükleme
+tokenizer = AutoTokenizer.from_pretrained(base_model_id, token=hf_token)
+base_model = AutoModelForCausalLM.from_pretrained(
+    base_model_id,
+    torch_dtype=torch.float32,
+    device_map=None,
+    token=hf_token
+)
+model = PeftModel.from_pretrained(base_model, lora_model_id, token=hf_token)
 model.eval()
 
+# 🚀 FastAPI app
 app = FastAPI()
 
 @app.post("/run/predict")